In this talk I will describe new algorithms and tools for generating paintings, illustrations, and animation using a computer. The long-term goal of this work is to combine the beauty and expressiveness of traditional media with the flexibility of computer graphics, with potential applications ranging from animated films to technical illustration.
I will begin with a painterly rendering algorithm for processing images and video sequences. This method produces images with a much greater subjective impression of “looking hand-made” than do earlier methods.
I will then present a new style of line art illustration for smooth 3D surfaces. This style is designed to clearly convey surface shape, even for surfaces without predefined material properties or hatching directions.
Finally, I will describe a new machine learning framework for processing images by example, called “image analogies.” Given an example of a painting or drawing (e.g. scanned from a hand-painted source), we can process new images with some approximation to the style of the painting, without requiring any explicit definition of the style. The image analogies framework supports many other novel image processing operations by example, such as “texture-by-numbers,” where we synthesize new realistic imagery from example photographs.